A lightning forecast and drone route recommendation method based on electric field data

Abstract

The present invention relates to the field of weather data processing technology, and specifically to a lightning forecast and drone route recommendation method based on electric field data, including outputting a first probability of lightning on each section of an inspection route; outputting a second probability of lightning on each section of an inspection route; fusing the first probability and the second probability in real time to obtain a fused third probability; judging whether the third probability of lightning on each section of the route exceeds a predetermined value, and when judging that the third probability of lightning on at least one section of the route exceeds a predetermined value, sending the regional position of the route section where the third probability exceeds the predetermined value to a drone control module, and planning the drone route based on the regional position by the drone control module to obtain a second route. The present invention can timely change the route according to the real-time predicted lightning situation, so that the drone can avoid the location with a high probability of lightning, thereby reducing the occurrence of the drone being struck by lightning.

Description

Technical Field

[0001] The present invention relates to the technical field of weather data processing, and in particular to a lightning forecasting and unmanned aerial vehicle route recommendation method based on electric field data. Background Art

[0002] Mountain lines usually pass through complex terrain environments, such as high mountains, deep valleys, dense forests, etc. These areas are often inconvenient for transportation and difficult for personnel to reach, which brings great challenges to line inspection. The traditional inspection method requires inspectors to walk through the mountains, which is not only time-consuming and labor-intensive, but also inefficient and has safety hazards. In order to improve the efficiency of line inspection, drones are currently used to replace human inspections. During the inspection, drones usually set the inspection route and then fly along the inspection route. However, in the event of extreme weather changes, such as lightning, drones cannot change the route in time according to the lightning situation. Summary of the invention

[0003] In view of this, the purpose of the present invention is to propose a lightning forecast and drone route recommendation method based on electric field data, so as to solve the problem that when there is lightning while the drone is conducting line inspection, the drone cannot change its route in time according to the lightning situation.

[0004] Based on the above purpose, the present invention provides a lightning forecasting and drone route recommendation method based on electric field data, comprising:

[0005] S1: Obtain the first prediction factor of the inspection line segment;

[0006] S2: input the obtained first prediction factor into the lightning prediction model for analysis and calculation, and output the first probability of lightning occurring on each section of the inspection line;

[0007] S3: Determine whether the first probability of lightning occurring on each section of the line exceeds a predetermined value. When it is determined that the first probability of lightning occurring on at least one section of the line exceeds the predetermined value, send the regional position of the line section where the first probability exceeds the predetermined value to the drone control module, and plan the drone route based on the regional position by the drone control module to obtain a first route.

[0008] S4: Controlling the flight of the UAV based on the first route, and obtaining in real time a second prediction factor measured by the UAV through an atmospheric electric field instrument on the UAV;

[0009] S5: input the obtained second prediction factor into the lightning prediction model for analysis and calculation, and output the second probability of lightning occurring on each section of the inspection line;

[0010] S6: fusing the first probability and the second probability in real time to obtain a fused third probability;

[0011] S7: determining whether the third probability of lightning occurring on each section of the route exceeds a predetermined value. When it is determined that the third probability of lightning occurring on at least one section of the route exceeds the predetermined value, sending the regional position of the route section where the third probability exceeds the predetermined value to the drone control module, and the drone control module plans the drone route based on the regional position to obtain a second route.

[0012] S8: Controlling the flight of the UAV based on the second route.

[0013] Optionally, the second route is a return route that avoids area locations of route segments where the third probability exceeds a predetermined value.

[0014] Optionally, the second route is a route that avoids the regional position of the line segment where the third probability exceeds a predetermined value, and then sequentially arrives at a safe distance outside the regional position of the line segment where the third probability exceeds a predetermined value. After reaching the safe distance, a real-time second prediction factor is obtained, and then the real-time third probability is obtained through fusion. Warning information is sent based on the real-time third probability, and the warning information includes the probability of lightning occurrence and preventive measures for power facilities.

[0015] Optionally, the second route is a route that avoids the regional position of the line segment where the third probability exceeds a predetermined value and reaches a safe distance outside the regional position of the line segment where the third probability exceeds the predetermined value the most. After reaching the safe distance, the lightning activity process is recorded by a high-resolution camera carried on the drone, and a real-time second prediction factor is obtained by the atmospheric electric field meter on the drone to track the location and intensity changes of lightning activities.

[0016] Optionally, the prediction factors include electric field intensity, the ratio of electric field intensity to sunny day electric field intensity, electric field intensity variability, variance, and whether the electric field undergoes polarity reversal.

[0017] Optionally, the first probability and the second probability are fused by weighted averaging.

[0018] Optionally, the lightning prediction model includes trained random forest units, logistic regression units, K-nearest neighbor units, Bayesian units, and support vector machine units. The probability of lightning occurring on each section of the inspection line segment of each unit is obtained by inputting the first prediction factor into the trained random forest units, logistic regression units, K-nearest neighbor units, Bayesian units, and support vector machine units, and the probability of lightning occurring on each section of the inspection line segment of each unit is fused to obtain the first probability.

[0019] When the UAV is conducting line inspection, it can obtain the first prediction factor of the inspection line segment through the atmospheric electric field meters arranged at intervals along the line, and then input the obtained first prediction factor into the lightning prediction model for analysis and calculation, and output the first probability of lightning occurring on each section of the inspection line segment; judge whether the first probability of lightning occurring on each section of the line exceeds a predetermined value. When it is judged that the first probability of lightning occurring on at least one section of the line exceeds the predetermined value, it indicates that there is a greater probability of lightning occurring on the section of the line. At this time, the regional position of the line segment where the first probability exceeds the predetermined value is sent to the UAV control module, and the UAV control module plans the UAV route based on the regional position to obtain the first route. The first route is a route that avoids the above-mentioned regional position, thereby avoiding the position with a greater probability of lightning occurring. Then, the flight of the UAV is controlled based on the first route, and the UAV is obtained in real time through the atmospheric electric field meter on the UAV. The second prediction factor obtained by the UAV measurement is input into the lightning prediction model for analysis and calculation, and the second probability of lightning occurring on each section of the inspection line is output. The first probability and the second probability are integrated in real time to obtain the integrated third probability; the third probability integrates the first probability obtained based on the relevant data measured by the ground atmospheric electric field instrument and the second probability obtained based on the relevant data measured by the atmospheric electric field instrument on the UAV, and can more accurately reflect the probability of lightning occurrence than the first probability. At this time, it is judged whether the third probability of lightning occurring on each section of the line exceeds the predetermined value. When it is judged that the third probability of lightning occurring on at least one section of the line exceeds the predetermined value, the regional position of the line section where the third probability exceeds the predetermined value is sent to the UAV control module, and the UAV control module plans the UAV route based on the regional position to obtain a second route, and controls the flight of the UAV based on the second route.

[0020] As can be seen from the above, the method first obtains the first probability through the relevant data measured by the ground atmospheric electric field meter. At this time, the atmospheric electric field meter on the drone is not working to save the power of the drone. When it is determined that the first probability of lightning on at least one section of the line exceeds the predetermined value, the original route is adjusted in time. The drone control module plans the drone route based on the regional position to obtain the first route, and then flies based on the first route to avoid the route of the above-mentioned regional position. At the same time, the atmospheric electric field meter on the drone is started, and the second probability is obtained through the relevant data measured by the atmospheric electric field meter on the drone. A more accurate third probability is obtained by fusing the first probability and the second probability, and then the latest regional position exceeding the predetermined value is obtained according to the third probability, and then the drone route is planned to obtain the second route. In this way, the route can be changed in time according to the real-time predicted lightning situation, so that the drone can avoid the location with a higher probability of lightning, thereby reducing the occurrence of the drone being struck by lightning. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 Flow chart of a method for lightning forecasting and drone route recommendation based on electric field data according to an embodiment of the present invention;

[0023] Figure 2 This is a diagram showing the importance of each prediction factor in the random forest (RF) method in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0025] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0026] like Figure 1 As shown, a lightning forecasting and drone route recommendation method based on electric field data includes:

[0027] S1: Obtain the first prediction factor of the inspection line segment;

[0028] S2: input the obtained first prediction factor into the lightning prediction model for analysis and calculation, and output the first probability of lightning occurring on each section of the inspection line;

[0029] S3: Determine whether the first probability of lightning occurring on each section of the line exceeds a predetermined value. When it is determined that the first probability of lightning occurring on at least one section of the line exceeds the predetermined value, send the regional position of the line section where the first probability exceeds the predetermined value to the drone control module, and plan the drone route based on the regional position by the drone control module to obtain a first route.

[0030] S4: Controlling the flight of the UAV based on the first route, and obtaining in real time a second prediction factor measured by the UAV through an atmospheric electric field instrument on the UAV;

[0031] S5: input the obtained second prediction factor into the lightning prediction model for analysis and calculation, and output the second probability of lightning occurring on each section of the inspection line;

[0032] S6: fusing the first probability and the second probability in real time to obtain a fused third probability. Optionally, the first probability and the second probability are fused by a weighted average method.

[0033] S7: determining whether the third probability of lightning occurring on each section of the route exceeds a predetermined value. When it is determined that the third probability of lightning occurring on at least one section of the route exceeds the predetermined value, sending the regional position of the route section where the third probability exceeds the predetermined value to the drone control module, and the drone control module plans the drone route based on the regional position to obtain a second route.

[0034] S8: Controlling the flight of the UAV based on the second route.

[0035] When the UAV is conducting line inspection, it can obtain the first prediction factor of the inspection line segment through the atmospheric electric field meters arranged at intervals along the line, and then input the obtained first prediction factor into the lightning prediction model for analysis and calculation, and output the first probability of lightning occurring on each section of the inspection line segment; set a predetermined value, for example, the predetermined value can be set to 50%, and judge whether the first probability of lightning occurring on each section of the line exceeds the predetermined value. When it is judged that the first probability of lightning occurring on at least one section of the line exceeds the predetermined value, it indicates that there is a greater probability of lightning occurring on the section of the line. At this time, the regional position of the line segment where the first probability exceeds the predetermined value is sent to the UAV control module, and the UAV control module plans the UAV route based on the regional position to obtain the first route. The first route is a route that avoids the above-mentioned regional position, thereby avoiding the position with a greater probability of lightning occurring. Then, the flight of the UAV is controlled based on the first route, and at the same time, the large The atmospheric electric field instrument obtains a second prediction factor obtained by the drone measurement in real time, inputs the obtained second prediction factor into the lightning prediction model for analysis and calculation, outputs a second probability of lightning occurring on each section of the inspection line segment, and fuses the first probability and the second probability in real time to obtain a fused third probability; the third probability fuses the first probability obtained based on the relevant data measured by the ground atmospheric electric field instrument and the second probability obtained based on the relevant data measured by the atmospheric electric field instrument on the drone, and can more accurately reflect the probability of lightning occurrence than the first probability. At this time, it is judged whether the third probability of lightning occurring on each section of the line exceeds a predetermined value. When it is judged that the third probability of lightning occurring on at least one section of the line exceeds the predetermined value, the regional position of the line segment where the third probability exceeds the predetermined value is sent to the drone control module, and the drone control module plans the drone route based on the regional position to obtain a second route, and controls the drone flight based on the second route.

[0036] As can be seen from the above, the method first obtains the first probability through the relevant data measured by the ground atmospheric electric field meter. At this time, the atmospheric electric field meter on the drone is not working to save the power of the drone. When it is determined that the first probability of lightning on at least one section of the line exceeds the predetermined value, the original route is adjusted in time. The drone control module plans the drone route based on the regional position to obtain the first route, and then flies based on the first route to avoid the route of the above-mentioned regional position. At the same time, the atmospheric electric field meter on the drone is started, and the second probability is obtained through the relevant data measured by the atmospheric electric field meter on the drone. A more accurate third probability is obtained by fusing the first probability and the second probability, and then the latest regional position exceeding the predetermined value is obtained according to the third probability, and then the drone route is planned to obtain the second route. In this way, the route can be changed in time according to the real-time predicted lightning situation, so that the drone can avoid the location with a higher probability of lightning, thereby reducing the occurrence of the drone being struck by lightning.

[0037] In some embodiments, the second route is a return route that avoids the area where the route segment with the third probability exceeds a predetermined value, so that the drone can return safely in the first place while avoiding the location with a higher probability of lightning.

[0038] In some embodiments, the second route is a route that avoids the regional position of the line segment where the third probability exceeds the predetermined value, and then sequentially arrives at a safe distance outside the regional position of the line segment where the third probability exceeds the predetermined value. After reaching the safe distance, a real-time second prediction factor is obtained, and then the real-time third probability is obtained through fusion. Warning information is sent based on the real-time third probability, and the warning information includes the probability of lightning occurrence and preventive measures for power facilities.

[0039] The second route drone in this embodiment can reach a route at a safe distance outside the regional location of the line segment where the third probability exceeds a predetermined value. Then the atmospheric electric field meter on the drone obtains the real-time second prediction factor at the location at a safe distance, and then obtains the real-time third probability through fusion. At this time, since the drone flies near the regional location where the third probability exceeds the predetermined value, the real-time second prediction factor obtained at this time can more accurately reflect the lightning situation at the regional location. Then the early warning system can send early warning information based on the real-time third probability. The early warning information includes the probability of lightning occurrence and the prevention measures of power facilities. The prevention measures of power facilities may include emergency preparations for lightning strikes.

[0040] In some cases, it is necessary to record the impact of lightning on the line. In order to better record the impact of lightning on the line, in some embodiments, the second route is a route that avoids the regional position of the line segment where the third probability exceeds the predetermined value, and then reaches a safe distance outside the regional position of the line segment where the third probability exceeds the predetermined value the most. After reaching the safe distance, a high-resolution camera mounted on the drone is used to record the lightning activity process and track the location and intensity changes of the lightning activity.

[0041] The second route in this embodiment allows the drone to reach a safe distance outside the regional location of the line segment where the third probability exceeds the maximum predetermined value. The regional location of the line segment where the third probability exceeds the maximum predetermined value is the location where lightning is most likely to occur. Therefore, after reaching the safe distance, the drone can more effectively record the lightning activity process through the high-resolution camera on the drone, so that it can be provided to relevant scientific researchers for subsequent use. At the same time, after arriving at the location, the real-time second prediction factor can be obtained through the atmospheric electric field meter on the drone, so as to provide advance warning for the next location of the lightning activity.

[0042] In some embodiments, the prediction factors include electric field intensity, ratio of electric field intensity to sunny day electric field intensity, electric field intensity variability, variance, and whether the electric field has polarity reversal. These factors not only take into account the sensitivity of different atmospheric electric field instruments and the systematic deviation caused by the installation environment, but also reflect several characteristics of the atmospheric electric field before and after the occurrence of lightning.

[0043] Table 1 Selected predictors

[0044]

[0045] In some embodiments, the lightning prediction model includes a trained random forest (RF) unit, a logistic regression (LR) unit, a K-nearest neighbor (KNN) unit, a Bayesian (GNB) unit, and a support vector machine (SVM) unit. By inputting a first prediction factor into the trained random forest (RF) unit, the logistic regression (LR) unit, the K-nearest neighbor (KNN) unit, the Bayesian (GNB) unit, and the support vector machine (SVM) unit, the probability of lightning occurring on each section of the inspection line segment of each unit is obtained, and the probability of lightning occurring on each section of the inspection line segment of each unit is fused to obtain a first probability.

[0046] Five methods, including random forest (RF), logistic regression (LR), K-nearest neighbor (KNN), Bayesian (GNB), and support vector machine (SVM), were used to train the samples. The parameters used by each method are shown in Table 2.

[0047] Table 2 Parameter settings of five machine learning classification methods

[0048]

[0049] The data set was randomly divided into a training set and a test set at a ratio of 3:1. Five machine learning methods were used to train the training set, and five prediction models were established and tested using the test set. From the test results (Table 3), it can be seen that the accuracy of each machine learning method is not much different. In comparison, the RF method has the highest accuracy of 0.780, followed by the LR and GNB methods, which are 0.778, and the KNN method has the lowest accuracy of 0.737. From the perspective of various indicators, the KNN method has the least missed reports and the most correctly predicted lightning times, while the LR method has the least false reports.

[0050] Table 3 Test results of lightning warning models using different machine learning algorithms using the test set

[0051]

[0052] Importance evaluation of each predictor

[0053] The RF algorithm can calculate the Gini coefficient as a measurement indicator based on the contribution of each prediction factor to each tree in the RF. The maximum contribution factor is defined as 100, and the relative values ​​of other factors and the maximum contribution factor are calculated as the importance of each prediction factor ( Figure 2 ). In the RF method, the electric field strength and whether polarity reversal occurs are the top two important factors, followed by the electric field strength variability, which can well reflect the atmospheric electric field fluctuations. The variance factor is ranked last, which has a certain similarity with the electric field strength variability. Different weights can be set for each prediction factor according to the importance of the prediction factor. The higher the importance, the greater the weight.

[0054] In this embodiment, the probability of lightning is calculated respectively by five units, namely, a random forest (RF) unit, a logistic regression (LR) unit, a K-nearest neighbor (KNN) unit, a Bayesian (GNB) unit and a support vector machine (SVM) unit, to obtain the probability of lightning occurring on each section of the inspection line segment of each unit, and the probability of lightning occurring on each section of the inspection line segment of each unit is fused to obtain the first probability. Compared with the calculation of a single unit, this can provide a more comprehensive and accurate calculation result.

[0055] It should be understood by those skilled in the art that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention (including the claims) is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the above aspects of the present invention, which are not provided in detail for the sake of simplicity.

[0056] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A lightning forecast and drone route recommendation method based on electric field data, characterized in that: include: S1: Obtain the first prediction factor of the inspection line segment; S2: Input the obtained first prediction factor into the lightning prediction model for analysis and calculation, and output the first probability of lightning occurring on each section of the inspection line; S3: Determine whether the first probability of lightning occurring on each section of the line exceeds a predetermined value. When it is determined that the first probability of lightning occurring on at least one section of the line exceeds the predetermined value, send the regional position of the line section where the first probability exceeds the predetermined value to the drone control module, and plan the drone route based on the regional position by the drone control module to obtain a first route. S4: Controlling the flight of the UAV based on the first route, and obtaining in real time a second prediction factor measured by the UAV through an atmospheric electric field instrument on the UAV; S5: input the obtained second prediction factor into the lightning prediction model for analysis and calculation, and output the second probability of lightning occurring on each section of the inspection line; S6: fusing the first probability and the second probability in real time to obtain a fused third probability; S7: Determine whether the third probability of lightning occurring on each section of the route exceeds a predetermined value. When it is determined that the third probability of lightning occurring on at least one section of the route exceeds the predetermined value, send the regional position of the route section where the third probability exceeds the predetermined value to the drone control module, and the drone control module plans the drone route based on the regional position to obtain a second route; S8: Controlling the flight of the UAV based on the second route; The second route is a route that avoids the regional position of the line segment where the third probability exceeds the predetermined value and reaches a safe distance outside the regional position of the line segment where the third probability exceeds the predetermined value the most. After reaching the safe distance, the lightning activity process is recorded by a high-resolution camera carried on the drone, and the real-time second prediction factor is obtained by the atmospheric electric field meter on the drone to track the location and intensity changes of lightning activities.

2. The lightning forecasting and drone route recommendation method based on electric field data according to claim 1 is characterized in that: The second route is a return route that avoids the area location of the route segment where the third probability exceeds a predetermined value.

3. The lightning forecasting and drone route recommendation method based on electric field data according to claim 1 is characterized in that: The second route is a route that avoids the regional position of the line segment where the third probability exceeds the predetermined value, and then sequentially arrives at a safe distance outside the regional position of the line segment where the third probability exceeds the predetermined value. After reaching the safe distance, a real-time second prediction factor is obtained, and then the real-time third probability is obtained through fusion. Warning information is sent based on the real-time third probability, and the warning information includes the probability of lightning occurrence and preventive measures for power facilities.

4. The lightning forecasting and drone route recommendation method based on electric field data according to claim 1 is characterized in that: The prediction factors include electric field intensity, the ratio of electric field intensity to sunny day electric field intensity, electric field intensity variability, variance, and whether the electric field has polarity reversal.

5. The lightning forecasting and drone route recommendation method based on electric field data according to claim 1 is characterized in that: The first probability and the second probability are fused by weighted averaging.

6. The lightning forecasting and drone route recommendation method based on electric field data according to claim 1 is characterized in that: The lightning prediction model includes trained random forest units, logistic regression units, K-nearest neighbor units, Bayesian units, and support vector machine units. The probability of lightning occurring on each section of the inspection line segment of each unit is obtained by inputting a first prediction factor into the trained random forest units, logistic regression units, K-nearest neighbor units, Bayesian units, and support vector machine units. The probability of lightning occurring on each section of the inspection line segment of each unit is fused to obtain a first probability.

Images